Skip to main content

AI-Assisted Development

The fastest way to build with PDFDancer is to let your AI coding assistant do it for you. Install the PDFDancer MCP Server once, then just describe what you want in plain English—your AI automatically searches the docs and implements the feature.

Installation​

One-Click Install​

Cursor:

Install in Cursor

Claude Code:

claude mcp add pdfdancer -- npx -y @pdfdancer/pdfdancer-mcp

Manual Installation​

For detailed installation instructions for all clients (Claude Code, VS Code, Windsurf, Zed, etc.), see the Installation Guide.

After installation, restart your AI assistant.

Requirements​

  • Node.js 18 or higher
  • MCP-compatible client (Claude Code, Cursor, VS Code with MCP extension, etc.)

How It Works​

  1. Install the MCP server (one-time setup, takes 3 seconds)
  2. Ask your AI: "Create a PDF invoice generator in Python, use pdfdancer"
  3. AI automatically:
    • Searches PDFDancer docs for relevant APIs
    • Finds code examples in your preferred language
    • Writes working code with proper error handling
    • Debugs issues by consulting the docs again

Real-World Example​

Here's a complete session where a developer went from zero to a working TypeScript project with custom image placement—all guided by the MCP server:

> create a typescript project which creates a pdf with the words
  "Hello World" on it. Use pdfdancer

⏺ I'll help you create a TypeScript project that generates a PDF
  with "Hello World" using PDFDancer. Let me start by exploring
  the PDFDancer documentation.

# AI searches documentation, finds quickstart guides, and:
# - Initializes npm project
# - Installs dependencies (typescript, pdfdancer-client-typescript)
# - Creates src/main.ts with working code
# - Sets up build scripts
# - Generates output.pdf

⏺ Done! The PDF is saved as output.pdf.

> add logo.png to the top-right corner

# AI searches for image documentation, then:
# - Creates a sample logo.png
# - Updates code to add the image
# - Handles API errors by consulting docs again
# - Rebuilds and generates updated PDF

⏺ Perfect! The PDF now contains the logo in the top-right
  and "Hello World" text.

Total time: ~2 minutes from idea to working application.

What You Can Build​

Once installed, just ask your AI in natural language. Here are real examples:

Document Processing​

You: "Create a function which takes a PDF file as input,
      finds the date on the first page and changes it to today,
      use pdfdancer"

→ AI searches docs, writes complete code with date parsing and replacement

Form Filling​

You: "Build a script that fills out the invoice template PDF
      with data from a CSV file, use pdfdancer"

→ AI finds AcroForm documentation, implements CSV parsing and form field population

Batch Processing​

You: "Create a TypeScript function that adds a confidential watermark
      to all pages of multiple PDFs in a folder, use pdfdancer"

→ AI implements file system iteration, watermark positioning, and batch processing

Text Extraction & Analysis​

You: "Write a Python script that extracts all phone numbers
      from a PDF contract and saves them to a JSON file, use pdfdancer"

→ AI combines text extraction with regex pattern matching and JSON serialization

Dynamic Report Generation​

You: "Create a Java function that generates a monthly sales report PDF
      with charts, a summary table, and company branding, use pdfdancer"

→ AI implements layout design, table generation, image placement, and text formatting

Document Modification​

You: "Build a function that finds all instances of a company name in a PDF
      and replaces it with a new name, preserving formatting, use pdfdancer"

→ AI searches for text selection APIs, implements find-and-replace with style preservation

Behind the scenes, your AI automatically searches the PDFDancer documentation, finds relevant APIs, and implements solutions. You just focus on describing what you want.

Troubleshooting​

AI Doesn't Seem to Know About PDFDancer​

Problem: Your AI doesn't automatically search PDFDancer docs when you ask.

Solutions:

  1. Verify Node.js 18+ is installed: node --version
  2. Check your config file exists at the correct path (see Installation Guide)
  3. Restart your AI assistant completely
  4. Try explicitly mentioning: "Create a PDF invoice, use pdfdancer"

Connection or Timeout Errors​

Problem: AI says it can't reach the documentation.

Solutions:

  1. Ensure you have internet connectivity
  2. Try running npx -y @pdfdancer/pdfdancer-mcp manually to test
  3. Check firewall settings aren't blocking npx
  4. Restart your AI assistant

AI Provides Outdated Code​

Problem: The code examples don't match current PDFDancer APIs.

Solutions:

  1. The MCP server always fetches the latest docs—this is likely a different issue
  2. Clear your AI assistant's cache/context and start a new session
  3. Explicitly say: "Use the latest PDFDancer SDK to..."

Try It Out​

Start with simple requests and build up complexity:

Beginner:

  • "Create a Python script that generates a PDF with 'Hello World', use pdfdancer"
  • "Add a page number at the bottom, use pdfdancer"

Intermediate:

  • "Build a TypeScript invoice generator with line items and totals, use pdfdancer"
  • "Extract all text from a PDF and save to a file, use pdfdancer"

Advanced:

  • "Create a Java application that merges multiple PDFs and adds a table of contents, use pdfdancer"
  • "Build a report generator with charts, images, and multi-column layouts, use pdfdancer"

Your AI handles everything—from project setup to implementation to debugging.

More Information​

Ready to build? Just ask your AI: "Create a PDF with..., use pdfdancer" and watch it implement the solution automatically.